Discrete bat algorithm and application in community detection

Abstract

With the rapid development of social media, users have more options to form spontaneously reading or watching communities, and detecting such communities has significance both in technological aspects and commercial aspects. Theoretically community detecting problem is NP-hard problem and thus people inclined to choose heuristic algorithms based on objective optimization. Among these algorithms Bat-inspired Algorithm (BA) was originally proposed to solve continuous objective optimization problems. In this paper, authors explore the role of Bat-inspired Algorithm on detecting community structure in networks. This paper firstly introduces the definition of community and Bat-inspired Algorithm, then gives Discrete Bat Algorithm (hereinafter referred to as D-BA), the detailed design of detecting a classic community structure in the Karate club network and the standard of evaluation. Finally analyzes and evaluates the result.